No abstract

This paper evaluates the relative importance of country and firm-specific determinants of capital structure using a multilevel modelling approach. Annual data for 18,201 public and non-financial firms from 66 countries are analysed for the period 2000-2016. Variance decomposition analysis is employed in order to assess the relative importance of country and firm levels. Additionally, random intercept and random coefficient models are used to analyse direct and indirect effects of capital structure determinants. Our results showed that country and firm levels explain approximately 10% and 60% of the total variability in capital structures, respectively. This shows that managers assign a higher importance to the firm-level factors when making capital structure decisions. Also country-level variables affect leverage choices to a lower extent.

As one of the most important models in the finance literature, the Capital Asset Pricing Model (CAPM) assumes the existence of a positive and linear relationship between the systematic risk and required rates of return on stocks. The model is extensively researched in the academia and frequently used in business world since its development half a century ago. Its popularity comes from the simplification it provides for the complex process of asset pricing by making the assumption that only one single factor affects stock returns. But, as this is an unrealistic assumption, the validity of the model in its standard (unconditional) form is repeatedly rejected by empirical tests. Pettengill et al., (1995) developed an alternative conditional CAPM approach where the standard model is improved by taking bull and bear market conditions into consideration. According to this model, there is a positive (negative) risk-return relationship during up (down) market periods. Using this reasoning, Pettengill et al., (1995) tested up and down market periods separately and reached highly significant results that support CAPM. In this study, both the unconditional and conditional versions of CAPM are tested in the Istanbul Stock Exchange (ISE) for the period of nine years from 2003 to 2011. The test period is divided into four sub-periods. The unconditional CAPM is rejected for the sample period. A result of the conditional test shows that there is a statistically significant conditional relationship during some sub-periods. However, since the risk-return relationship in up and down markets is not symmetric, this conditional relationship does not indicate a positive risk-return tradeoff. Thus, CAPM may not be a useful asset pricing model for the ISE.

Bu çalışmada gelişmekte olan ülkelerde sermaye yapısı belirleyicileri Hiyerarşik Lineer Model (HLM) yaklaşımı kullanılarak analiz edilmiştir. Oluşturulan panel veri seti ülke, sektör, firma ve zaman düzeylerinden oluşan hiyerarşik bir yapı olarak düşünülmüş ve kaldıraç oranındaki değişkenlik varyans bileşenleri modeli ile incelenmiştir. Çalışma sonucunda, gelişmekte olan ülkelerdeki firmaların kaldıraç oranlarındaki değişkenliğin büyük kısmının firma ve zaman düzeylerinden kaynaklandığı gözlemlenmiştir. Ülkeye ve firmaya özgü faktörlerin sermaye yapısı üzerindeki doğrudan ve dolaylı etkilerini incelemek için ise rassal sabit ve rassal eğim modelleri kullanılmıştır. Gelişmekte olan ülkelerde, firmaya özgü faktörlerin halka açık firmaların sermaye yapısı kararları üzerinde ülkeye özgü faktörlere kıyasla çok daha etkili olduğu bulgulanmıştır. Ülkeye özgü faktörlerin ise sermaye yapısı üzerindeki doğrudan etkilerinin yanında firmaya özgü sermaye yapısı belirleyicileri üzerinden dolaylı etkilerinin de olduğu söylenebilir.

This panel data set comprises of the annual factoring payables and financial ratios of 261 publicly listed firms in Turkey for the period of 2012–2017. The raw data are drawn from three different sources; the Central Bank of Republic of Turkey (CBRT) database, the audit reports disclosed through the Public Disclosure Platform (KAP) of Turkey and Compustat Global Capital IQ (Compustat) database. The originality of this data set is the revealment of factoring payables at the firm level in Turkey for the first time. This paper and its dataset are a companion for a published article in the Borsa Istanbul Review under the title “Factoring as a determinant of capital structure for large firms: Theoretical and empirical analysis” [1].